Thesis (M.S.)--University of Texas at El Paso, 2009. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Constraint programming is an invaluable tool for solving many of the complex NP-complete problems that we need solutions to. These problems can be easily described as Constraint Satisfaction Problems (CSPs) and then passed to constraint solvers: complex pieces of software written to solve general CSPs efficiently. Many of the problems we need solutions to are real world problems: planning (e.g. vehicle routing), scheduling (e.g. job shop schedules) and timetabling problems (e.g. staff rotas) to name but a few. In the real world, we place structure on objects to make them easier to deal with. This manifests itself as symmetry. The symmetry in these real world problems make them easier to deal with for humans. However, they lead to a great deal of redundancy when using computational methods of problem solving. Thus, this thesis examines some of the many aspects of utilising the symmetry of CSPs to reduce the amount of computation needed by constraint solvers. In this thesis we look at the ease of use of previous symmetry breaking methods. We introduce a new and novel method of describing the symmetries of CSPs. We look at previous methods of symmetry breaking and show how we can drastically reduce their computation while still breaking all symmetry. We give the first detailed investigation into the behaviour of breaking only subsets of all symmetry. We look at how this affects the performance of constraint solvers before discovering the properties of a good symmetry. We then present an original method for choosing the best symmetries to use. Finally, we look at areas of redundant computation in constraint solvers that no other research has examined. New ways of dealing with this redundancy are proposed with results of an example implementation which improves efficiency by several orders of magnitude.

Aujourd'hui, les éditeurs logiciels ne conçoivent, développent et ne maintiennent plus leur offre logicielle avec comme cible un client unique. Au contraire, les offres logicielles sont conçues pour cibler plusieurs entités. Par conséquent, ces applications doivent s'intégrer dans des environnements différents et s'adapter aux besoins des clients. Ainsi, les produits logiciels développés ne sont plus des programmes uniques, mais des familles de produits. Les systèmes configurables facilitent la création de ces familles de produits. Grâce à eux il est possible de créer un produit logiciel en sélectionnant les fonctionnalités qui seront intégrées. Cependant, la validation de ces systèmes  est une tâche complexe. Un système configurable peut générer plusieurs millions de configurations possibles. Il ne s'agit donc plus de valider un seul et unique produit, mais un ensemble de produits. Cet important nombre de configurations est un problème pour les personnes chargées de la validation. Nous proposons trois contributions qui visent à mieux répondre aux problématiques liées à la variabilité lors des projets de test : une présentation détaillée de deux projets de test industriels faisant face à des problématiques de variabilité issus de deux entreprises : Cisco et Orange ; une méthode originale basée sur les techniques de programmation par contraintes pour extraire des configurations de test qui respectent le critère Pairwise à partir d'un modèle explicite de la variabilité ; une comparaison de cette approche par rapport aux techniques de l'état de l'art et une étude de l'application de cette technique de test sur deux projets de tests industriels. / Nowadays, software companies develop and maintain their software for several clients. Consequently, these applications have to  be integrated in heterogenous context and adapt to the user requriements. All these products are sharing commonalities but also differ in certain point due to business specific constraints. Configurable systems facilitate the creation of these product families. With them it is possible to create a software product by selecting the features that will be integrated, thus, the creation of a product is greatly simplified. However,  the validation of these systems is a complex task. A configurable system can generate millions of possible configurations. Thus, validation process doesn't consist in validating a single product but in validating a set of products. This large number of configurations is a problem for those responsible of the validation. In this thesis we propose three contributions that aim to solve issues raised by  variability during test projects : A detailled presentation of two industrial test projects coping tat variaibility issues; an original methodology based on constraint programming techniques to select test configurations that respect pairwise criteria from a feature model ; an exhaustive comparison of this approach with the existing approches and a detailled study of the application of a such techniques on the two industrials projects.

Invoice processing has traditionally been heavily dependent onmanual labor, where the task is to identify and move certaininformation from an origin to a destination. A time demandingtask with a high interest of automation to reduce time ofexecution, fault-risk and cost.With the evergrowing interest in automation and ArtificialIntelligence (AI), this thesis will explore the possibilities ofautomating the task of extracting and mapping information ofinterest by defining the problem as a Constraint OptimizationProblem (COP) using numeric relations between present information.The problem is then solved by extracting the numericalvalues in a document and utilizing it as an input space whereeach combination of numeric values are tested using a backendsolver.Several different models were defined, using different approachesand constraints on relations between possible existingfields. A solution to an invoice was considered correct if thetotal, tax, net and rounding amounts were estimated correctly.The final best achieved results were 84.30% correct and8.77% incorrect solutions on a set of 1400 various types of invoices.The achieved results show a promising alternative route toproposed solutions using e.g. machine learning or other intelligentsolutions using graphical or positional data. While only regardingthe numerical values present in each document, the proposedsolution becomes decentralized and therefor can be implementedand ran on any set of invoices without any pre-training phase.

Cette thèse présente de nouvelles techniques pour les Constraint Games.La manière de résoudre un Constraint Game est repensée en terme de propagation de contraintes.Les préférences des joueurs sont maintenant considérées comme des contraintes globales permettant une intégration transparente dans les solveurs de contraintes ainsi que d'améliorer l'efficacité du framework.Notre nouveau solveur ConGA est diffusé en open source.Celui-ci est plus rapide que les travaux connexes et est capable de trouver tous les équilibres de Nash, et cela même dans des jeux avec 200 joueurs voir 2000 pour certains jeux graphiques.Grâce à cette perspective, le framework a pu être utilisé pour résoudre un problème de routage dans le domaine des télécommunications. Les aspects centralisé et décentralisé ont été étudiés.La comparaison de ces derniers est très importante pour évaluer la qualité de service dans les applications multi-utilisateurs. L'évaluation de cette dernière peut être très coûteuse, c'est pourquoi nous proposons plusieurs techniques permettant d'améliorer la résolution de ce problème et ainsi d'améliorer la résolution du problème. / This thesis revisits the Constraint games framework by rethinking their solving technique in terms of constraint propagation.Players preferences are considered as global constraints making transparently the integration in constraints solvers.It yields not only a more elegant but also a more efficient framework.We release our new solver ConGA in open source.Our new complete solver is faster than previous state-of-the-art and is able to find all pure Nash equilibrium for some problems with 200 players or even with 2000 players in graphical games.This new perspective enables us to tackle real-worlds Telecommunication problems.This problem is solved with a centralized perspective and a decentralized one.The comparison of the two last approaches is really important to evaluate the quality of service in multi-users application, but computationally consuming.That is why, we propose new techniques in order to improve the resolution process.

Les diagrammes de décision Multi-valués (MDD) sont des structures de données efficaces et largement utilisées dans les domaines tels que la vérification, l’optimisation et la programmation dynamique. Dans cette thèse, nous commençons par améliorer les principaux algorithmes tels que la réduction de MDD, permettant aux MDD de potentiellement compresser exponentiellement des ensembles de tuples, ou la combinaison de MDD, tels que l’intersection ou l’union. Ensuite, nous proposons des versions parallèles de ces algorithmes ainsi que des versions permettant de travailler avec la version non déterministe des MDD. De plus, dans le domaine des MDD relâchés, un domaine de plus en plus étudié, nous définissons les notions de réduction et combinaison relâchés, ainsi que leurs algorithmes associés. Nous résolvons le problème de l’échantillonnage des solutions d’un MDD avec respect de loi de probabilité tels que des fonctions de probabilité de masse ou des chaines de Markov. Pour permettre d’utiliser les MDD dans les solveurs de programmation par contraintes, nous proposons de nouveaux propagateurs pour toutes les contraintes basées sur des MDD, améliorant les performances des algorithmes existants, puis nous en introduisons une nouvelle contrainte, la contrainte de channeling. Grâce à eux, nous montrons que nous pouvons reformuler plusieurs contraintes et en définir de nouvelles tout en étant basés sur des MDD. Finalement nous appliquons nos algorithmes à des problèmes industriels réels de génération de texte et musique, et de modélisation de réservoir de pétrole. / Multivalued Decision Diagrams (MDDs) are efficient data structures widely used in several fields like verification, optimization and dynamic programming. In this thesis, we first focus on improving the main algorithms such as the reduction, allowing MDDs to potentially exponentially compress set of tuples, or the combination of MDDs such as the intersection of the union. We go further by designing parallel algorithms, and algorithms handling non-deterministic MDDs. We then investigate relaxed MDDs, that are more and more used in optimization, and define the notions of relaxed reduction or operation and design efficient algorithms for them. The sampling of solutions stored in a MDD is solved with respect to probability mass functions or Markov chains. In order to combine MDDs with constraint Programming, we design the propagators of all the types of MMDD constraints in solvers, and introduce a new one, the channeling constraint. These new propagators outperform the existing ones and allow the reformulation of several other constraints such as the dispersion constraint, and even to define new ones easily. We finally apply our algorithm to several real world industrial problems such as text and music generation and geomodeling of a petroleum reservoir.

This thesis presents an approach for modeling and generating efficient hardware architectures using constraint programming techniques, targeting field programmable gate arrays (FPGAs). The focus of this thesis is the derivation of optimal or near-optimal schedules for streaming applications from data flow graphs (DFGs). The resulting schedules are then used to facilitate the architecture generation process. Most streaming applications, like digital singal processing (DSP) algorithms, are repetitive in nature: the same computation is performed on different data items. This repetitive nature of streaming applications can be used to expose additional parallelism available across different iterations, by creating multiple instances of the same computation. The replication of the single computation, when applied to high level synthesis (HLS), improves the performance of the design but requires additional area. The amount of additional area required for a replicated graph can be reduced through the use of pipelined functional units and the addition of some extra clock cycles beyond the critical path of the DFG. This thesis discusses the use of a constraint programming (CP)-based scheduler to generate optimal schedules based on designer-provided replication level and critical path relaxation. The scheduler is an integrated part of the design tool, called CHARGER, which analyzes the resulting schedules to allocate memory for storing intermediate data, creates the infrastructure necessary to efficiently execute the application, and finally generates a synthesizable Verilog/VHDL code for the controller. The performance of the architectures derived using the CP-based scheduler is compared with the architectures generated using a force directed scheduling (FDS)-based scheduler for algorithms selected from embedded/multimedia applications. The results show that our CP-based scheduler outperforms the FDS-based scheduler, both in terms of area and efficiency of the generated architectures. The results show average area saving of 39% and average performance improvement of 41%.

No description available.

This thesis explores three new techniques for increasing the efficiency of constraint propagation: support for incremental propagation, improved representation of constraints, and abstractions to simplify propagation.  Support for incremental propagation is added to a propagator centered propagation system by adding a new intermediate layer of abstraction, advisors, that capture the essential aspects of a variable centered system. Advisors are used to give propagators a detailed view of the dynamic changes between propagator runs. Advisors enable the implementation of optimal algorithms for important constraints such as extensional constraints and Boolean linear in-equations, which is not possible in a propagator centered system lacking advisors.  Using Multivalued Decision Diagrams (MDD) as the representation for extensional constraints is shown to be useful for several reasons. Classical operations on MDDs can be used to optimize the representation, and thus speeding up the propagation. In particular, the reduction operation is stronger than the use of DFA minimization for the regular constraint. The use of MDDs is contrasted and compared to a recent proposal where tables are compressed.  Abstractions for constraint programs try to capture small and essential features of a model. These features may be much cheaper to propagate than the unabstracted program. The potential for abstraction is explored using several examples. These three techniques work on different levels. Support for incremental propagation is essential for the efficient implementation of some constraints, so that the algorithms have the right complexity. On a higher level, the question of representation looks at what a propagator should use for propagation. Finally, the question of abstraction can potentially look at several propagators, to find cases where abstractions might be fruitful. An essential feature of this thesis is a novel model for general placement constraints that uses regular expressions. The model is very versatile and can be used for several different kinds of placement problems. The model applied to the classic pentominoes puzzle will be used through-out the thesis as an example and for experiments. / Den här avhandlingen utforskar tre nya tekniker för att öka effektiviteten av villkorspropagering: stöd för inkrementell propagering, val av representation för villkor, samt abstraktion för att förenkla propagering. Ett propageringssystem organiserat efter propagerare utökas med stöd för inkrementell propagering genom att lägga till ett nytt abstraktionslager: rådgivare. Detta lager fångar de essentiella aspekterna hos system organiserade efter variabler. Rådgivare används för att ge propagerare detaljerad information om de dynamiska ändringarna i variabler mellan körningar av propageraren. Utökningen innebär att det går att implementera optimala algoritmer för vissa viktiga villkor såsom tabellvillkor och Boolska linjära olikheter, något som inte är möjligt i ett rent propagator-organiserat system. Användandet av så kallade Multivalued Decision Diagram (MDD) som representation för tabellvillkor visas vara användbart i flera avseenden. Klassiska MDD-operationer kan användas för att optimera representationen, vilket leder till snabbare propagering. Specifikt så är reduktionsoperationen kraftfullare än användandet av DFA-minimering för reguljära villkor. MDD-representationen jämförs också med ett nyligen framlagt förslag för komprimerade tabeller. Abstraktioner för villkorsprogram försöker fånga små men viktiga egenskaper i modeller. Sådana egenskaper kan vara mycket enklare att propagera än den konkreta modellen. Potentialen för abstraktioner undersöks för några exempel. Dessa tre tekniker fungerar på olika nivåer. Stöd för inkrementell propagering är nödvändigt för att kunna implementera vissa villkor effektivt med rätt komplexitet. Valet av representation för villkor är på en högre nivå, då det gäller att se vilka algoritmer som skall användas för ett villkor. Slutligen så måste flera villkor i en modell studeras för att finna rätt typ av abstraktioner. Ett utmärkande drag för den här avhandlingen är en ny modell för generella placeringsvillkor som använder reguljära uttryck. Modellen är mångsidig och kan användas för flera olika typer av placeringsproblem. Modellen specialiserad för pentominopussel används genomgående som exempel för experiment. / QC 20101117 / Coordinating Constraint Propagation

Constraint Programming (CP) is a powerful technique for solving large-scale combinatorial (optimisation) problems. Constraint solving a given problem proceeds in two phases: modelling and solving. Effective modelling has an huge impact on the performance of the solving process. This thesis presents a framework in which the users are not required to make modelling decisions, concrete CP models are automatically generated from a high level problem specification. In this framework, modelling decisions are encoded as generic rewrite rules applicable to many different problems. First, modelling decisions are divided into two broad categories. This categorisation guides the automation of each kind of modelling decision and also leads us to the architecture of the automated modelling tool. Second, a domain-specific declarative rewrite rule language is introduced. Thanks to the rule language, automated modelling transformations and the core system are decoupled. The rule language greatly increases the extensibility and maintainability of the rewrite rules database. The database of rules represents the modelling knowledge acquired after analysis of expert models. This database must be easily extensible to best benefit from the active research on constraint modelling. Third, the automated modelling system Conjure is implemented as a realisation of these ideas; having an implementation enables empirical testing of the quality of generated models. The ease with which rewrite rules can be encoded to produce good models is shown. Furthermore, thanks to the generality of the system, one needs to add a very small number of rules to encode many transformations. Finally, the work is evaluated by comparing the generated models to expert models found in the literature for a wide variety of benchmark problems. This evaluation confirms the hypothesis that expert models can be automatically generated starting from high level problem specifications. An method of automatically identifying good models is also presented. In summary, this thesis presents a framework to enable the automatic generation of efficient constraint models from problem specifications. It provides a pleasant environment for both problem owners and modelling experts. Problem owners are presented with a fully automated constraint solution process, once they have a precise description of their problem. Modelling experts can now encode their precious modelling expertise as rewrite rules instead of merely modelling a single problem; resulting in reusable constraint modelling knowledge.